The creators of the recent AI phenomenon ChatGPT have co-authored a research paper examining the impact of large-scale language models (LLMs) on the labor market. The paper reveals the impact of the generative pre-trained transformer (GPT) model on a variety of specific occupations and even provides a list of occupations where the impact is zero.

In other words, if you're lucky, talented, and dedicated enough to be a professional athlete, your job is safe until you give AI not only language, but legs. Huh? That means AI already has legs.

In all, 34 different occupations are listed as unaffected by LLM - this is the index of the paper on the impact of LLM.

The paper titled "GPT is GPT: An Early Look at the Labor Market Impact Potential of Large Language Models (opens in new tab)" (PDF warning enclosed in graph) is OpenAI, OpenResearch, and the University of Pennsylvania. Frankly, given how much skin the authors have in the LLM game, one would have to take the findings more or less with a grain of salt.

However, if you want to have the confidence to resist the AI revolution, I suggest you retrain for one of these occupations:

They are all fairly self-evident and based mostly on skilled physical labor. That said, I am a bit concerned about the amount of effort that went into this study, since it says that the occupation of "dishwasher" seems to be fine. I mean, let's face it, AI is unlikely to replace human dishwashers.

But if we don't want future law masters to have a terrible time, we need to stay out of professions like journalism. Oh.

According to this paper, 19% of professions will have at least 50% of their jobs exposed by ChatGPT and its ilk. But that does not necessarily mean that their jobs will be replaced, but potentially augmented or subsidized.

Indeed, Sam Altman, CEO of OpenAI, has publicly stated that AI models are to be used as a tool, not a complete replacement for humans.

"AI waits for someone to input," Altman said in an interview with ABC (opens in new tab). 'This is a very human-controlled tool.'

This is a view shared by Google, and I asked Byrd if he should worry about his own ChatGPT rival and his work as a technology journalist.

Overall, Google's answer was very positive and spoke to the growing demand for technology journalists as more and more people turn to technology to stay informed and connected. He also gave me a little advice about developing sources, being concise, and not writing sloppy sentences, which I tend to do.

"Thank you. What about the impact of LLMs like yourself?" [LLMs like me can have a huge impact on technology journalism. For example, they can be used to generate large amounts of content, such as news articles, blog posts, product reviews, etc. This way, journalists can focus on more in-depth reporting and analysis.

In addition, LLMs can be used to translate content into different languages, thus reaching a wider audience. It can also be used to generate personalized content, such as news articles tailored to readers' interests.

Overall, LLMs like mine can be a valuable tool for technology journalists. It is important to note, however, that I am not a substitute for a human journalist. I can only do what I am trained to do, and I cannot replace the human touch that is essential to good journalism."

These are pleasant words. Nevertheless, my biggest concern I have with LLM is how confidently it can lie and misinform. Even on something as seemingly simple as the number of words in a sentence. So I still don't know if I should take career advice from Bard or ChatGPT.

My chances of becoming a professional athlete are long gone. I don't have the guts to work in a slaughterhouse.